Subscriber access provided by Nottingham Trent University
Environmental Modeling
Mapping the annual flow of steel in the United States Yongxian Zhu, Kyle Syndergaard, and Daniel R Cooper Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.9b01016 • Publication Date (Web): 30 Aug 2019 Downloaded from pubs.acs.org on August 30, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 27
Environmental Science & Technology
U.S. steel flow | Revision 1
1
Mapping the annual flow of steel in the
2
United States
3 4
Yongxian Zhu1, Kyle Syndergaard1, Daniel R Cooper1* 1. Mechanical Engineering Department, University of Michigan,
5
George G. Brown Laboratory,
6
2350 Hayward Street, Ann Arbor, MI, USA, 48109-2125
7 8
*Corresponding
author phone: (734) 764-1357, email address:
[email protected] 9 10
Abstract art
11 12 13
ACS Paragon Plus Environment
1
Environmental Science & Technology
Page 2 of 27
U.S. steel flow | Revision 1
14
Abstract
15
A detailed understanding of material flows is needed to target increased
16
material efficiency and the circular economy. In this article, the U.S. steel
17
flow is modeled as a series of nodes representing processes and products.
18
An easily updatable nonlinear least squares optimization is used to
19
reconcile the inconsistencies across 293 collated data records on flows
20
through and between the nodes. The data come from an integrated
21
analysis that includes top-down estimates of steel flow from trade bodies
22
and government statistical agencies, bottom-up estimates of the steel
23
embedded in products based on production statistics and bills of materials,
24
and the mass of imports and exports based on international money flows.
25
A weighting methodology is used to consistently assign confidence scores
26
to the data and the optimization is used to achieve mass balance and
27
minimize the sum of the squares of the weighted residuals. The results
28
indicate that yield improvement efforts should focus on sheet metal
29
forming in the car industry, which accounts for nearly half of all generated
30
fabrication scrap. The quantity of end-of-life scrap exported and landfilled
31
is greater than the quantity of steel products imported. Increased
32
domestic recycling of end-of-life scrap might displace around a third of
33
these imports.
34
Keywords: circular economy, material efficiency, material flow analysis,
35
data reconciliation, nonlinear optimization
36
ACS Paragon Plus Environment
2
Page 3 of 27
Environmental Science & Technology
U.S. steel flow | Revision 1
37
Introduction
38
The steel industry accounts for 30% of global industrial greenhouse gas
39
emissions (GHG) 1. The Intergovernmental Panel on Climate Change
40
(IPCC) recommends an overall 40% to 70% reduction in GHG emissions
41
from 2010 levels by 2050 2. However, with current best steelmaking
42
practices already approaching thermodynamic limits, even deployment of
43
cutting-edge production technologies will not be enough for the steel
44
industry to meet the IPCC’s emissions targets 3,4.
45
The realization that steel production must decrease if emissions targets
46
are to be achieved has helped lead to new research areas under the
47
banners of ‘material efficiency’ 5 and the ‘circular economy’ 6, both aimed
48
at reducing emissions-intensive material production. Researchers in these
49
new areas require a detailed material map in order to identify
50
opportunities.
51
Unlike in the developing world, U.S. per capita steel stocks plateaued
52
around 1980. The stock saturation level has been estimated at 9.1-14.3
53
t/capita 7–9. Per capita stocks are expected to saturate in much of the
54
developing world to a level similar to those in the U.S. by the late 21st
55
century 10,11. Therefore, the derived U.S. consumption pattern may
56
represent a population-scaled surrogate model of the future global state.
57
Previous steel maps and production statistics
58
A detailed snapshot of global production and consumption in 2008 is
59
provided by Cullen et al. 12, and Pauliuk et al. 9 estimate the in-use iron
60
stocks for 200 countries for the same year. Wang et al. construct global
61
and country level iron cycles for the year 2000 13. Other global flows focus
62
on the production of crude steel without analyzing the flow of intermediate
63
products7,14,15.
ACS Paragon Plus Environment
3
Environmental Science & Technology
Page 4 of 27
U.S. steel flow | Revision 1 64
There have been numerous studies on steel use in regions and states,
65
including for the U.K. 16,17, Japan and Asia 18,19, the U.S.7–9,20, and North
66
America 10. However, these studies either present steel flow data at such a
67
low resolution as to make it difficult to glean detailed recommendations or
68
are mainly concerned with steel stock levels and scrap discards, which are
69
only a portion of the overall steel flow.
70
U.S. focused studies that provide a one year snapshot of the steel flow
71
include Andersen and Hyman, who create calibrated energy and material
72
flow models for the steel industry in 1994 based on publicly available data
73
and starting with raw materials and proceeding through to semi-finished
74
products 21. Müller et al. constructed a flow diagram for steel in 2000
75
including imports and exports, but only showing aggregated flows of
76
products (e.g., “construction”) 8.
77
The World Steel Association (WSA) releases a yearly dataset showing
78
production, consumption, and trade data for over 80 countries 22. High
79
resolution domestic production data are presented for intermediate
80
products such as hot rolled coil or construction reinforcement bar (rebar).
81
The WSA also publishes international trade data as mass flows but
82
aggregates direct trade (imports and exports of steelmaking raw materials
83
such as iron ore and steel mill products such as cold rolled coil) into broad
84
categories such as “flat products,” and only provide an overall indirect
85
trade mass flow. Indirect trade is of finished products (e.g., automobiles)
86
that contain steel. The WSA provides no information on finished goods
87
fabrication or scrap generation and trade.
88
The United States Geological Survey (USGS), using data largely derived
89
from the trade body American Iron and Steel Institute 23, presents more
90
specifics on the U.S. steel industry than the WSA. A yearly “Minerals
91
Yearbook” has sections on iron and steel 24, iron and steel scrap 25, and
92
iron ore 26. Unlike the WSA, USGS reports granular data on intermediate
93
product (direct) imports and exports and scrap consumption. Neither
ACS Paragon Plus Environment
4
Page 5 of 27
Environmental Science & Technology
U.S. steel flow | Revision 1 94
USGS nor the WSA publish statistics with standard deviation errors;
95
however, errors are clearly present that manifest themselves as
96
discrepancies both within and across the data sources. One source of error
97
is the process of collecting and aggregating the data through regular
98
surveys of steel companies. For example, the Iron and Steel Scrap section
99
of the 2015 USGS Minerals Yearbook 25 notes that data are derived from
100
voluntary monthly or annual surveys, and that about 68% of known pig
101
iron and raw steel producers responded that year, representing only 32%
102
of the total scrap consumed that year. USGS reports data for the most
103
recent year and several previous years. Numbers for previous years have
104
often been revised as the result of continuing industry survey returns.
105
There is therefore a tradeoff between the pertinence and the reliability of
106
the data when using the comprehensive USGS datasets to help examine
107
steel flows.
108
Scope of this article
109
A detailed U.S. steel material flow analysis (MFA) is needed to determine
110
the production (and hence emissions) attributable to U.S. consumption
111
and to identify the most effective strategies to reduce steel demand. This
112
article focuses on 2014 as the most recent year for which detailed and
113
reliable production and intermediate product data are available from
114
USGS and the WSA.
115
The MFA is tabulated in the Supporting Information (Table S2) and is
116
presented in the main article as a Sankey diagram, which is a common
117
form of depicting energy and mass flows 27,28. The flow from U.S. mining
118
and scrap purchases to final U.S. consumption (flow of steel into use) is
119
shown sequentially from the left side of the diagram to the right. The
120
width of the lines on the diagram are proportional to the size of the mass
121
flows.
ACS Paragon Plus Environment
5
Environmental Science & Technology
Page 6 of 27
U.S. steel flow | Revision 1
122
Methodology
123
Several methods exist which could generate a U.S. steel map. Economic
124
data could be used to assign steel flows to monetary flows based on
125
commerce reporting. However, formal input-output tables only provide
126
sectoral level resolution (e.g., construction) and the conversion from money
127
flows to steel flows varies widely among products. Otherwise, top-down data
128
on steel production (e.g., the WSA Statistical Yearbook) can be used to
129
estimate low resolution steel flows to the level of intermediate products
130
(e.g., wire rod) and in some cases low resolution sectors (e.g., transport). The
131
opposite approach of using bottom-up data is based on combining sales data
132
for specific classes of products with average bills of materials.
133
An integrated analysis is used in this article that leverages the knowledge
134
embedded in all the above techniques. Data from trade organizations (e.g.,
135
the WSA), governmental scientific agencies (e.g., USGS), and academic
136
literature (e.g., Wang et al. 13) is combined with monetary trade statistics
137
(e.g., Comtrade data on imports and exports 29), and bottom-up estimates
138
derived by the authors (see Data records on U.S. manufacturing…). The
139
steel flow is modeled as a series of connected nodes representing major
140
steel processing technologies (e.g., the blast furnace) and major products
141
used and created by industry (e.g., iron ore or passenger cars). Data
142
records from the integrated analysis are catalogued (S3) under the
143
corresponding flow coordinate shown in Figure 1. For example, USGS 26
144
states that 26.8 Mt of iron (contained within iron ore) enters the blast
145
furnace (BF). This datum is catalogued under the coordinate (1,2).
146
Multiple, potentially conflicting data records may be catalogued under the
147
same coordinate; e.g., data sources report that domestically produced pig
148
iron exports in 2014 were equal to 6.77 kt (USGS 24), 7 kt (USGS: (31)),
149
and 52 kt (WorldSteel 22). All these records are catalogued under the
150
coordinate (2,49). Unconventional data referring to multiple flows are
151
catalogued as well. For example, USGS 24 does not record the production
ACS Paragon Plus Environment
6
Page 7 of 27
Environmental Science & Technology
U.S. steel flow | Revision 1 152
of continuously cast billets, blooms, and slabs separately (each is a “node”
153
in the steel map) but does record the sum of the three. This datum is
154
recorded under coordinate (8,55).
155
Inconsistencies between the collated and derived data records are
156
reconciled using a least squares optimization model (see Data
157
reconciliation). The next three sections describe the modeled steel flow and
158
the origin of key data records used to generate the steel map. All data
159
records used in this analysis can be found in the Supporting Information
160
(SI).
161 162 163
Figure 1: The coordinate system used to define the steel flow and catalogue data records. See S1.2 for complete details of the cataloguing method
164
Data records on the U.S. steel industry (Nodes: 1-23)
165
Steel “flows” from liquid steel production through casting, intermediate
166
product manufacturing, fabrication of end-use goods, use, and finally end-
167
of-life (EOL) processing. There are additional flows into and out of these
168
categories in the form of imports, metal losses, scrap generation, and
169
exports.
ACS Paragon Plus Environment
7
Environmental Science & Technology
Page 8 of 27
U.S. steel flow | Revision 1
170
Liquid metal production
171
In primary steelmaking, iron ore is first converted to pig iron in a blast
172
furnace (BF) and then to steel in a basic oxygen furnace (BOF). Small
173
amounts of scrap, DRI, and other iron inputs are consumed alongside iron
174
ore in the BF. The quantities of each input are reported by USGS 24. Iron
175
ore production, import, and export are reported in the USGS Minerals
176
Yearbook in both the Iron and Steel section and the Iron Ore section 24,26.
177
Iron ore is also used to produce direct reduced iron (DRI), which is another
178
raw material used in steelmaking. The production, import, and export of
179
DRI are reported in the WSA Yearbook 22 and by Midrex 31.
180
Liquid high-carbon-content pig iron is typically sent straight from the BF
181
to the BOF, where it is combined with scrap that helps to cool the melt 15.
182
However, the pig iron may instead be cast into ingots that are later used
183
for iron castings, scrap contaminant dilution in EAFs, or in BOFs not
184
situated near the BF. The WSA reports the production, import, and export
185
of pig iron in the U.S. 22.
186
In secondary steelmaking, scrap metal is melted in electric arc furnaces
187
(EAFs) 24. EOL and manufacturing scrap consumption is reported in the
188
USGS Minerals Yearbook in the Iron and Steel Scrap section 25. Some pig
189
iron and DRI, up to 50% of the melt 32, is also used in the EAF to control
190
the concentration of steel scrap impurities.
191
A small percentage (50% of industry 3:
